Material tag recommendation method and device, electronic equipment and storage medium

Abstract

This disclosure relates to a method, apparatus, electronic device, and storage medium for recommending content tags. The method includes: in response to a content tag recommendation request from a target terminal, acquiring and sending a list of content tag combinations to the target terminal; the list of content tag combinations is obtained based on a tag combination recommendation model that performs tag combination prediction processing on the content tags and display association information corresponding to multiple historical content; the display indicator information corresponding to the content tag combinations in the content tag combination list meets the display indicator conditions; in response to a filtering request for the content tag combination list, extracting target object information from the filtering request; acquiring a sub-list of content tag combinations matching the target object information from the content tag combination list; and sending the sub-list of content tag combinations to the target terminal. According to the technical solution provided by this disclosure, accurate content tags can be recommended for content creation, improving content creation efficiency and resource utilization.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer application, and particularly relates to a material label recommendation method and device, electronic equipment and storage medium. BACKGROUND

[0002] At present, in order to facilitate material producers to produce materials, material production tools are concerned, such as short video production tools. The material producers can conveniently produce materials in the material production tools and push the materials to a video library, but the material production tools generally only provide convenient processing of panels and editing, and do not give relevant processing information for content expression of material production, resulting in low efficiency of material production and low resource utilization rate. SUMMARY

[0003] The present disclosure provides a material label recommendation method, device, electronic equipment and storage medium. The technical solution of the present disclosure is as follows:

[0004] According to a first aspect of an embodiment of the present disclosure, a material label recommendation method is provided, comprising:

[0005] In response to a material label recommendation request of a target terminal, a material label combination list is acquired and sent to the target terminal, the material label combination list being obtained by label combination prediction processing of material labels and display associated information corresponding to a plurality of historical materials based on a label combination recommendation model; display index information corresponding to a material label combination in the material label combination list satisfies display index conditions;

[0006] In response to a filtering request for the material label combination list, target object information is extracted from the filtering request, the target object information being used to indicate associated information of a target object to which a material is pushed;

[0007] A material label combination sub-list matched with the target object information is acquired from the material label combination list;

[0008] The material label combination sub-list is sent to the target terminal.

[0009] In a possible implementation manner, the method further comprises:

[0010] Periodically acquiring a plurality of historical materials and display associated information corresponding to each historical material;

[0011] Inputting the plurality of historical materials into a material label prediction model to perform material label prediction processing, to obtain material labels corresponding to each historical material;

[0012] Input the material identifier corresponding to each historical material, the display associated information, and the material label into the label combination recommendation model, perform label combination recommendation processing, and obtain the material label combination list;

[0013] The material label corresponding to each historical material is at least one of a preset material label set, and the material labels in the material label combination are multiple in the preset material label set. A preset material label in the preset material label set represents information of a material in multiple production dimensions.

[0014] In a possible implementation, the method further includes:

[0015] Upon receiving a detail request for a target material label combination from the target terminal, obtain target detail page associated information corresponding to the target material label combination. The target material label combination is any material label combination in the material label combination list.

[0016] Send the target detail page associated information to the target terminal.

[0017] In a possible implementation, the method further includes:

[0018] In response to a matching test request from the target detail page, obtain a target material to be matched.

[0019] Perform prediction processing on the material label of the target material based on a material label prediction model, and obtain material label information corresponding to the target material.

[0020] Perform matching processing on the target material label combination and the material label information, and obtain a matching result of the target material and the target material label combination.

[0021] In a possible implementation, the matching processing on the target material label combination and the material label information to obtain the matching result of the target material and the target material label combination includes:

[0022] Determine a first label that is the same in the target material label combination and the material label information and a second label after merging and deduplication.

[0023] According to first weighting information of the first label and second weighting information of the second label, determine similarity information of the target material label combination and the material label information.

[0024] Based on the similarity information, obtain the matching result.

[0025] In a possible implementation, the method further includes:

[0026] in the case that the matching result meets the matching condition, sending the matching result to the target terminal and pushing the target material to a material library;

[0027] in the case that the matching result does not meet the matching condition, obtaining material tag analysis information and sending the material tag analysis information and the matching result to the target terminal.

[0028] In a possible implementation, the obtaining of the target detail page associated information corresponding to the target material tag combination comprises:

[0029] obtaining detailed description information corresponding to each material tag in the target material tag combination and reference material corresponding to the target material tag combination;

[0030] obtaining target detail page associated information corresponding to the target material tag combination according to the detailed description information and the reference material.

[0031] In a possible implementation, the obtaining and sending of the material tag combination list to the target terminal in response to a material tag recommendation request of the target terminal comprises:

[0032] obtaining the material tag combination list, preset object information corresponding to each material tag combination in the material tag combination list, recommendation attribute information and a corresponding reference video list in response to a material tag recommendation request of the target terminal;

[0033] generating material tag combination recommendation information according to the material tag combination list, the preset object information, the recommendation attribute information and the reference video list;

[0034] sending the material tag recommendation information to the target terminal.

[0035] According to a second aspect of the embodiments of the present disclosure, a material tag recommendation method is provided, comprising:

[0036] displaying a material tag combination recommendation page in response to a material tag recommendation request, the material tag combination recommendation page displaying a material tag combination list, the material tag combination list being obtained by label combination prediction processing on material tags and display associated information corresponding to a plurality of historical materials respectively based on a label combination recommendation model; display index information corresponding to a material tag combination in the material tag combination list meets display index conditions;

[0037] switching to display a material tag combination sub-list in the material tag combination recommendation page in response to a filtering request on the material tag combination list.

[0038] In a possible implementation, the method further includes:

[0039] In response to a request for details of a target material tag combination, a target detail page corresponding to the target material tag combination is displayed, and the target detail page displays target detail page associated information corresponding to the target material tag combination; the target material tag combination is any material tag combination in the material tag combination list.

[0040] In a possible implementation, the target detail page further displays a matching test control, and the method further includes:

[0041] In response to a matching test request triggered based on the matching test control, a target material to be matched is obtained.

[0042] The matching result of the target material and the target material tag combination is received and displayed.

[0043] According to a third aspect of the embodiments of the present disclosure, a material tag recommendation device is provided, including:

[0044] A material tag combination list recommendation module is configured to perform, in response to a material tag recommendation request of a target terminal, obtaining and sending a material tag combination list to the target terminal, the material tag combination list being obtained by performing label combination prediction processing on material tags corresponding to a plurality of historical materials and display associated information based on a label combination recommendation model; the display index information corresponding to the material tag combination in the material tag combination list meets the display index condition.

[0045] A material tag combination list screening module is configured to perform, in response to a screening request for the material tag combination list, extracting target object information from the screening request, the target object information being used to indicate the associated information of a target object to which a material is pushed.

[0046] A sub-list obtaining module is configured to perform obtaining a material tag combination sub-list matched with the target object information from the material tag combination list.

[0047] A sub-list recommendation module is configured to perform sending the material tag combination sub-list to the target terminal.

[0048] In a possible implementation, the device further includes:

[0049] An obtaining module is configured to perform periodically obtaining a plurality of historical materials and display associated information corresponding to each historical material.

[0050] The first prediction module is configured to perform material tag prediction processing on the plurality of historical material input material tag prediction models to obtain material tags corresponding to each historical material.

[0051] The second prediction module is configured to perform label combination recommendation processing on the material identifier corresponding to each historical material, the display association information, and the material tag input into the label combination recommendation model to obtain the material tag combination list.

[0052] The material tag corresponding to each historical material is at least one of a preset material tag set, and the material tags in the material tag combination are multiple in the preset material tag set. The preset material tag in the preset material tag set represents information of a material in multiple production dimensions.

[0053] In a possible implementation, the apparatus further includes:

[0054] The detail response module is configured to perform the following: receiving a detail request for a target material tag combination from the target terminal, and obtaining target detail page association information corresponding to the target material tag combination. The target material tag combination is any material tag combination in the material tag combination list.

[0055] The sending module is configured to perform the following: sending the target detail page association information to the target terminal.

[0056] In a possible implementation, the apparatus further includes:

[0057] The matching test response module is configured to perform the following: in response to a matching test request from the target detail page, obtaining a target material to be matched.

[0058] The third prediction module is configured to perform the following: based on a material tag prediction model, performing prediction processing on a material tag of the target material to obtain material tag information corresponding to the target material.

[0059] The matching test module is configured to perform the following: performing matching processing on the target material tag combination and the material tag information to obtain a matching result of the target material and the target material tag combination.

[0060] In a possible implementation, the matching test module includes:

[0061] The label determination unit is configured to perform the following: determining a first label that is the same in the target material tag combination and the material tag information, and a second label after merging and deduplication.

[0062] The similarity information determination unit is configured to determine similarity information of the target material tag combination and the material tag information according to first weighting information of the first tag and second weighting information of the second tag.

[0063] The matching result acquisition unit is configured to acquire the matching result based on the similarity information.

[0064] In a possible implementation, the apparatus further includes:

[0065] The first matching processing module is configured to send the matching result to the target terminal and push the target material to a material library if the matching result meets a matching condition.

[0066] The second matching processing module is configured to acquire material tag analysis information and send the material tag analysis information and the matching result to the target terminal if the matching result does not meet the matching condition.

[0067] In a possible implementation, the detail response module includes:

[0068] The description information and reference material acquisition unit is configured to acquire detailed description information corresponding to each material tag in the target material tag combination and reference material corresponding to the target material tag combination.

[0069] The detail association information acquisition unit is configured to acquire target detail page association information corresponding to the target material tag combination according to the detailed description information and the reference material.

[0070] In a possible implementation, the material tag combination list recommendation module includes:

[0071] The tag recommendation response unit is configured to acquire the material tag combination list, preset object information corresponding to each material tag combination in the material tag combination list, recommendation attribute information, and a corresponding reference video list in response to a material tag recommendation request of the target terminal.

[0072] The recommendation information generation unit is configured to generate material tag combination recommendation information according to the material tag combination list, the preset object information, the recommendation attribute information, and the reference video list.

[0073] The recommendation information sending unit is configured to send the material tag recommendation information to the target terminal.

[0074] According to a fourth aspect of the embodiments of the present disclosure, a material tag recommendation apparatus is provided, including:

[0075] The recommendation page display module is configured to display a material tag combination recommendation page in response to a material tag recommendation request, the material tag combination recommendation page displaying a material tag combination list obtained by performing label combination prediction processing on material tags corresponding to a plurality of historical materials and display association information based on a label combination recommendation model; and the display index information corresponding to the material tag combinations in the material tag combination list satisfies display index conditions.

[0076] The recommendation switching display module is configured to switch to display a material tag combination sub-list in the material tag combination recommendation page in response to a screening request for the material tag combination list.

[0077] In a possible implementation, the apparatus further includes:

[0078] The detail page display module is configured to display a target detail page corresponding to a target material tag combination in response to a detail request for the target material tag combination, the target detail page displaying target detail page association information corresponding to the target material tag combination; and the target material tag combination is any material tag combination in the material tag combination list.

[0079] In a possible implementation, the target detail page further displays a matching test control, and the apparatus further includes:

[0080] The matching test module is configured to display a matching result of a target material and the target material tag combination in response to a matching test request triggered based on the matching test control; and the target material is a material corresponding to the matching test request.

[0081] According to a fifth aspect of the embodiments of the present disclosure, an electronic device is provided, including: a processor; a memory for storing instructions executable by the processor; and wherein the processor is configured to execute the instructions to implement the method according to any one of the first aspect.

[0082] According to a sixth aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, when instructions in the computer readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the method according to any one of the first aspect of the embodiments of the present disclosure.

[0083] According to a seventh aspect of the embodiments of the present disclosure, a computer program product is provided, including computer instructions, when the computer instructions are executed by a processor, the computer is enabled to perform the method according to any one of the first aspect of the embodiments of the present disclosure.

[0084] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0085] By setting the material tag recommendation request and the screening request for the material tag combination list, the recommendation of the material tag combination list meeting the display index condition can be realized, and the associated information of the target object to be pushed can be screened in the material tag combination list, so that more accurate recommendation of the material tag combination sub-list is realized, and the convenience and flexibility of the material tag combination recommendation are improved. In addition, the material tag combination list is obtained by label combination prediction processing of the respective material tags of the plurality of historical materials and the display associated information based on the label combination recommendation model, which can improve the accuracy of the material tag combination recommendation and the recommendation timeliness when facing a large amount of historical materials. In addition, the display index information corresponding to the material tag combination in the material tag combination list meets the display index condition, so that the material made according to the material tag combination in the material tag combination list can obtain a better display effect. Based on these, the cost of material making can be effectively reduced, and the resource utilization rate and the efficiency of material making can be improved.

[0086] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0087] The accompanying drawings incorporated in the specification and forming a part of it, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure without imposing undue limitation on the disclosure.

[0088] Figure 1 is a schematic diagram of an application environment according to an exemplary embodiment.

[0089] Figure 2 is a flowchart of a method for obtaining a material tag combination list according to an exemplary embodiment.

[0090] Figure 3 is a flowchart of a material tag recommendation method according to an exemplary embodiment.

[0091] Figure 4a is a schematic diagram of a material tag combination recommendation page according to an exemplary embodiment.

[0092] Figure 4b is a schematic diagram of a material tag combination recommendation page according to an exemplary embodiment.

[0093] Figure 5 is a schematic diagram of a material tag combination detail page according to an exemplary embodiment.

[0094] Figure 6 is a method flowchart of a material matching test according to an example embodiment.

[0095] Figure 7 is a display schematic diagram of a matching result according to an example embodiment.

[0096] Figure 8 is an interaction schematic diagram of a material tag combination recommendation according to an example embodiment.

[0097] Figure 9 is a material tag recommendation device block diagram according to an example embodiment.

[0098] Figure 10 is a block diagram of an electronic device for material tag recommendation according to an example embodiment.

[0099] Figure 11 is a block diagram of an electronic device for material tag recommendation according to an example embodiment. DETAILED DESCRIPTION

[0100] In order to make the ordinary person skilled in the art better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below in conjunction with the drawings.

[0101] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation described in the following example embodiments does not represent all implementations consistent with the present disclosure. Rather, they are only examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0102] Artificial intelligence (AI) is the use of digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology and machine learning / deep learning and other directions.

[0103] In recent years, with the research and progress of artificial intelligence technology, artificial intelligence technology has been widely applied in many fields. The scheme provided in the embodiments of the present application relates to machine learning / deep learning technology, and is specifically explained as follows:

[0104] Please refer to Figure 1 , Figure 1 is a schematic diagram of an application environment according to an exemplary embodiment, as Figure 1 shown, the application environment can include a server 01 and a terminal 02.

[0105] In an optional embodiment, the server 01 can be used for material tag recommendation processing. Specifically, the server 01 can be a standalone physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Basic cloud computing services such as platform.

[0106] In an optional embodiment, the terminal 02 can be used to request a recommended material tag combination list, filter material tag combinations, and request a matching test, etc., as a reference for material production. Specifically, the terminal 02 can include, but is not limited to, smart phones, desktop computers, tablet computers, notebook computers, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, smart wearable devices, and other types of electronic devices. Optionally, the operating system running on the electronic device can include, but is not limited to, the Android system, the IOS system, Linux, Windows, etc.

[0107] In addition, it should be noted that Figure 1 shown is only one application environment of the material tag recommendation method provided by the present disclosure.

[0108] In the embodiments of the present application, the server 01 and the terminal 02 described above can be connected directly or indirectly through wired or wireless communication, which is not limited in the present application.

[0109] It should be noted that the steps shown in the drawings are only one possible order, and in fact they do not have to be strictly followed. Some steps can be executed in parallel without being dependent on each other. The user-related information and data involved in the present disclosure (including but not limited to data for display, training data, etc.) are all information and data authorized by the user or fully authorized by all parties.

[0110] Figure 2 is a method flowchart for obtaining a material tag combination list according to an exemplary embodiment. In order to provide a reference for material producers to improve the recommendation effect of materials, a material tag combination list with a better recommendation effect is selected periodically. Based on this, the process of obtaining the material tag combination list can include:

[0111] In step S201, a plurality of historical materials and the display association information corresponding to each historical material are periodically obtained.

[0112] In the embodiments of the present specification, the display association information can refer to the relevant information of each historical material being displayed within a preset period of time, such as the display amount, the association information of the display corresponding object, etc. The association information can include the industry, the object's preferred material type, etc., which are not limited by the present disclosure. Among them, the object can refer to an account, and accordingly, the display corresponding object can refer to the account to which at least one of the plurality of historical materials has been displayed.

[0113] In a possible implementation manner, a plurality of historical materials can be periodically obtained from a material library, for example, the materials added to the material library within a preset period of time can be obtained as a plurality of historical materials for the prediction processing of the material tag combination list. Accordingly, the display association information corresponding to each historical material within the preset period of time can be obtained. Among them, the preset period of time can be set based on the timeliness requirement, which is not limited by the present disclosure. As an example, the length of the preset period of time can be the same as the period corresponding to the periodicity, for example, the current time is t2, the time corresponding to the previous period is t1-t2, and the period T=t2-t1. Accordingly, the preset period of time can be a period of t1-t2, T length, that is, the historical materials and display association information between t1-t2 can be used to predict the material tag combination list with a better recommendation effect, to be used as a reference for material production within the subsequent t2-t3 period. Among them, t3-t2=T.

[0114] In step S203, the plurality of historical materials are input into a material tag prediction model for material tag prediction processing, to obtain the material tag corresponding to each historical material.

[0115] In the embodiments of this specification, multiple historical materials are input into a material tag prediction model for material tag prediction processing to obtain material tags corresponding to each historical material. Each material tag can be at least one of a preset material tag set. These preset material tags represent information about the material across multiple production dimensions, such as content production dimension, visual production dimension, and audiovisual production dimension. Each production dimension can include at least one preset material tag. These preset material tags can be pre-set according to actual needs, and this disclosure does not limit this. The preset material tags can be set to multiple levels. As an example, such as... Figure 4a and Figure 4b As shown, there can be two levels. For example, first-level material tags can include AAA, BBB, CCC, and DDD, etc. Correspondingly, AAA-1 to AAA-3 and AAA-5 can be second-level material tags under AAA; BBB-1 and BBB-2 can be second-level material tags under BBB; CCC-1 to CCC-4 can be second-level material tags under CCC; and DDD-1 can be a second-level material tag under DDD. This is just an example. This disclosure does not limit the content and number of first-level material tags, nor the content and number of second-level material tags under each first-level material tag.

[0116] In one example, the primary material tags under the content creation dimension could include character design; the primary material tags under the visual creation dimension could include scene aesthetics, image clarity, etc.; and the primary material tags under the audiovisual production dimension could include audiovisual production. Taking audiovisual production as an example, the corresponding secondary material tags could include vertical full-width layout, frame with end, etc. This disclosure does not impose any limitations on these.

[0117] The material label prediction model can be obtained by periodically training a pre-defined machine learning model based on a massive amount of sample material. This disclosure does not limit the specific training process. For example, the pre-defined machine learning model can be trained in a supervised manner using the labeled material tags from a massive amount of sample material. In each iteration, gradient backpropagation based on loss information is used to update the parameters of the pre-defined machine learning model until the iteration conditions are met. The pre-defined machine learning model that meets the iteration conditions can then be used as the material label prediction model. As an example, the iteration conditions could be a threshold for the number of iterations, loss information less than a threshold, etc., which this disclosure does not limit. The massive amount of sample material can be material that existed before a predetermined time and within a predetermined duration. For example, if the period of the material label prediction model is greater than T, then the predetermined time can be t1, and the predetermined duration can be greater than T. This disclosure does not limit this; when performing material label prediction, only the latest updated material label prediction model needs to be obtained.

[0118] In step S205, the material identifier, display association information, and material tags corresponding to each historical material are input into the tag combination recommendation model for tag combination recommendation processing to obtain a list of material tag combinations; the material tags in the material tag combination can be multiple preset material tags from a preset material tag set.

[0119] In the embodiments of this specification, the material identifiers, display association information, and material tags corresponding to each historical material can be input into a tag combination recommendation model for tag combination recommendation processing to obtain a list of material tag combinations. The tag combination recommendation model can be used to predict material tag combinations to be recommended, thus forming the material tag combination list. For example, by inputting the material identifiers, display association information, and material tags corresponding to each historical material into the tag combination recommendation model for tag combination recommendation processing, multiple material tag combinations whose display indicator information meets the display indicator conditions can be obtained, and these multiple material tag combinations can be combined into a material tag combination list. These multiple material tag combinations can be any combination of material tags from historical materials. That is, the tag combination recommendation model can predict the display association information corresponding to different material tag combinations. The display association information can include display indicator information and the association information of the corresponding objects. Based on this, the tag combination recommendation model can output the multiple material tag combinations whose display indicator information meets the display indicator conditions, as well as the display association information corresponding to each of the multiple material tag combinations. Further, the material tag combination list can be stored for subsequent material tag combination recommendations. Optionally, the material tag combination list and the display association information corresponding to each of the multiple material tag combinations in the material tag combination list can be stored.

[0120] The tag combination recommendation model can be obtained by periodically training a preset deep learning model based on the material identifiers, display association information, and corresponding material tags of the sample materials. As an example, this period can be the same as the period of the preset machine learning model mentioned above, where the material tags corresponding to the sample materials can be obtained based on the material tag prediction model of the previous period. For example, a sample material set can be obtained, such as sample materials within a predetermined time period before the current training period of the tag combination recommendation model, forming a sample material set. This allows the acquisition of the material identifiers, display association information, and corresponding material tags for each sample material in the sample material set. These material identifiers, display association information, and corresponding material tags can then be input into the preset deep learning model to learn the embedding vectors of each material tag and the display association information corresponding to different material tag combinations. Each iteration of learning can output multiple material tag combinations and corresponding display association prediction information. The display association prediction information can include predicted display metric information and the predicted association information of the objects corresponding to the displays. This approach optimizes the predicted display metrics to meet display metric conditions, thereby adjusting the parameters of a pre-defined deep learning model. For example, a loss is determined based on the predicted display metrics and the display metric conditions to adjust the model's parameters until the difference between the predicted display metrics and the display metric conditions is less than a specified threshold. This yields a tag combination recommendation model. The training process of this model can be viewed as a learning process targeting the display metric conditions, thereby learning material tag combinations that satisfy those conditions. The display metric conditions can refer to a threshold for the number of impressions. Optionally, the display metrics and display metric conditions can be set for the cold start phase; this disclosure does not limit this.

[0121] By periodically predicting the list of material tag combinations using a tag combination recommendation model, a list of material tag combinations for recommendation can be automatically obtained, which can improve the efficiency and accuracy of material tag recommendation. Furthermore, when faced with massive amounts of material, it can efficiently predict the list of material tag combinations that meet the display index conditions, effectively meeting the timeliness requirements of material tag recommendation.

[0122] Figure 3 This is a flowchart illustrating a material tag recommendation method according to an exemplary embodiment. For example... Figure 3 As shown, it may include the following steps.

[0123] In step S301, in response to the target terminal's material tag recommendation request, a list of material tag combinations is obtained and sent to the target terminal.

[0124] The list of material tag combinations can be obtained by performing tag combination prediction processing on the material tags and display association information corresponding to multiple historical materials based on a tag combination recommendation model. The display indicator information corresponding to the material tag combinations in the list meets the display indicator conditions. The display indicator information corresponding to the material tag combinations can refer to the indicator information that materials created based on material tag combinations can be displayed in material recommendations, such as the number of times they can be displayed in material recommendations. Correspondingly, the display indicator conditions can refer to the number of times they can be displayed.

[0125] In the embodiments described in this specification, the user on the target terminal side triggers entry. Figure 4a When the page shown is accessed, a material tag recommendation request can be generated and sent to the server. In response, the server retrieves a list of material tag combinations from its storage, which is the most recent period (within the current time frame), and sends this list to the target terminal, allowing the target terminal to display the material tag combination list on the material tag combination page. The most recent period (within the current time frame) refers to the period between the generation time of the material tag combination list and the current time where the time difference is minimal.

[0126] In one possible implementation, in addition to sending a list of material tag combinations, other information corresponding to each material tag combination in the list can also be sent, such as preset object information, recommendation attribute information, and a corresponding list of reference videos. Based on this, in response to a material tag recommendation request from the target terminal, the list of material tag combinations, the preset object information corresponding to each material tag combination in the list, the recommendation attribute information, and the corresponding list of reference videos can be obtained. The preset object information can refer to the material tag combination being preferred within the preset object corresponding to that preset object; that is, materials created based on that material tag combination are suitable for being pushed to that preset object. As an example, the preset object information can be the object's associated information, such as object information, the object's preferred material type, etc., and the object information can include industries, etc. The recommendation attribute information can characterize the degree of recommendation of materials under the material tag combination in the material tag combination list, such as high, medium, or low. The list of reference videos corresponding to the material tag combination can refer to videos that match that material tag combination; that is, the reference videos in the list possess all the material tags in that material tag combination and can serve as examples of video production for that material tag combination.

[0127] Furthermore, based on the list of material tag combinations, preset object information, recommended attribute information, and a list of reference videos, material tag combination recommendation information can be generated; and this recommendation information can be sent to the target terminal. Correspondingly, the material tag recommendation information displayed on the material tag combination recommendation page of the target terminal can be as follows: Figure 4aAs shown, this displays three material tag combinations, along with the corresponding preset object information, recommended attribute information, and a list of reference materials for each material tag combination. For example, the preset object information can be as follows: Figure 4a The filter conditions shown are: Unrestricted, Scenario Drama; Filter condition 2, Scenario Drama; Unrestricted, Food. The serial number can be negatively correlated with the recommended attribute information, allowing material tag combinations with higher recommended attribute information to be displayed first. It should be noted that the material tag combination recommendation page can include more than 3 material tag combinations; therefore, users can swipe the page to view more recommended material tag combinations. By generating and sending material tag combination recommendation information to the target terminal based on the material tag combination list, preset object information, recommended attribute information, and reference video list, the target terminal can display richer information about each material tag combination in the material tag combination list, making it easier for material creators to quickly understand any material tag combination and improve their production references.

[0128] In step S303, in response to a filtering request for the list of material tag combinations, target object information is extracted from the filtering request. This target object information is used to indicate the associated information of the target object to which the material is pushed, and may be one or more of the preset object information.

[0129] In the embodiments described in this specification, such as Figure 4a As shown, the material tag combination recommendation page can also include a filtering area, which can be used to filter related information of different objects, such as different object information, material types, etc. Figure 4a The selection boxes for Filter Condition 1, Filter Condition 2, and Filter Condition 3 are used to filter different object information. For example, Filter Condition 1 can be "Industry," while Filter Conditions 2 and 3 can be "Other Object Information." When the source material is video, the video type can include all videos, sitcoms, food videos, etc. Figure 4a As shown, the video type is all videos.

[0130] In practical applications, content creators can further filter content tag combinations on the content tag combination recommendation page to find combinations that match the target audience they want to push. Based on this, they can select from a selection box to submit a filtering request; for example, they can press Enter after making a selection, or the content tag combination recommendation page can include a submit button to trigger the filtering request. The target device can send a filtering request to the server, and the server, in response to the filtering request for the content tag combination list, can extract the target audience information from the filtering request. For example, in a sitcom / drama.

[0131] In step S305, a sub-list of material tag combinations that match the target object information is obtained from the material tag combination list;

[0132] In step S307, the material tag combination sublist is sent to the target terminal.

[0133] In the embodiments of this specification, a sub-list of material tag combinations matching the target object information can be obtained from the material tag combination list. For example, the target object information can be matched with preset object information, thereby obtaining the material tag combinations corresponding to the preset object information matching the target object information from the material tag combination list, which serves as the sub-list of material tag combinations. Furthermore, the sub-list of material tag combinations can be sent to the target terminal. Taking a scenario drama as an example, the display of the material tag combination recommendation page can be switched to... Figure 4b As shown, the preset object information corresponding to the material tag combinations is all situational drama.

[0134] By setting up material tag recommendation requests and filtering requests for the material tag combination list, it is possible to recommend material tag combinations that meet display metric conditions, and to filter the relevant information of the target objects to be pushed according to needs within the material tag combination list, achieving more accurate recommendations of sub-lists of material tag combinations and improving the convenience and flexibility of material tag combination recommendations. Furthermore, the material tag combination list is obtained by using a tag combination recommendation model to predict the combinations of material tags and display association information corresponding to multiple historical materials, which can improve the accuracy of material tag combination recommendations and the timeliness of recommendations when dealing with massive amounts of historical materials. In addition, if the display metric information corresponding to the material tag combinations in the material tag combination list meets the display metric conditions, materials created based on the material tag combinations in the list can achieve better display effects. Based on these factors, the cost of material production can be effectively reduced, and the resource utilization and efficiency of material production can be improved.

[0135] Figure 5 This is a schematic diagram illustrating a material tag combination details page according to an exemplary embodiment. Based on this, it can be triggered... Figure 4a The method allows users to access a specific combination of material tags to navigate to its details page for a more comprehensive understanding. Accordingly, the method may further include: receiving a details request from a target terminal for a target material tag combination; obtaining the associated information for a target details page corresponding to the target material tag combination; the target material tag combination can be any combination of material tags from the list of material tag combinations; and then sending the associated information for the target details page to the target terminal so that the target terminal can display information such as... Figure 5 The material tag combination details page shown is the details guide page, which can then display the associated information of the target details page.

[0136] As an example, retrieving the target details page association information corresponding to the target material tag combination can include: retrieving the detailed description information corresponding to each material tag in the target material tag combination (such as...). Figure 5 The example shown is 501), along with reference materials corresponding to the target material tag combination; and based on the detailed description information and reference materials, the associated information of the target details page corresponding to the target material tag combination can be obtained. Here, reference materials can be such as... Figure 5 The reference video shown.

[0137] By requesting details of the target material tag combination, you can obtain the associated information of the target detail page corresponding to the target material tag combination, so that the material creator can fully understand the target material tag combination and provide a more comprehensive production reference.

[0138] Figure 6 This is a flowchart illustrating a method for material matching testing according to an exemplary embodiment. Figure 6 As shown, in one possible implementation, the method may further include:

[0139] In step S601, in response to a matching test request from the target details page, the target material to be matched is obtained.

[0140] In the embodiments described in this specification, in response to a matching test request from a target details page, the target material to be matched can be extracted from the matching test request.

[0141] In step S603, the material tags of the target material are predicted based on the material tag prediction model to obtain the material tag information corresponding to the target material;

[0142] In step S605, the target material tag combination and material tag information are matched to obtain the matching result of the target material and the target material tag combination.

[0143] In one example, step S605 may include: determining the first tag that is the same as the target material tag combination and the second tag after merging and deduplication; determining the similarity information between the target material tag combination and the material tag information based on the first weighted information of the first tag and the second weighted information of the second tag; thereby obtaining the matching result based on the similarity information. The target material tag combination may be carried in the matching test request or determined based on the target details page, such as determining the target material tag combination corresponding to the target details page. By matching the target material tag combination and the material tag information using the weights of each material tag, the matching result between the target material and the target material tag combination can be obtained, reflecting the weight of different material tags in the matching process and improving the flexibility and accuracy of the matching result.

[0144] For example, suppose the material tag information is [t2,t3,t4], and the target material tag combination is [t3,t4,t5,t6]. The weights corresponding to t2,t3,t4,t5,t6 are w2,w3,w4,w5,w6. Then the similarity information can be (w3+w4) / (w2+w3+w4+w5+w6), so the similarity information can be directly used as the matching result; or the correspondence between similarity information and matching results can be obtained, and the matching result can be a quantitative matching result, such as high matching degree, medium matching degree, and low matching degree. Furthermore, based on this correspondence, the matching result matching the similarity information can be obtained, such as high matching degree. Here, t3,t4 can refer to the first tag, and t2,t3,t4,t5,t6 can refer to the second tag; (w3+w4) can refer to the first weighted information, and (w2+w3+w4+w5+w6) can refer to the second weighted information.

[0145] Optionally, if the matching result meets the matching conditions, the matching result can be sent to the target terminal, and the target material can be pushed to the material library. Optionally, pushing to the material library can also be triggered by the creator, and this disclosure does not limit this.

[0146] If the matching result does not meet the matching conditions, material tag analysis information can be obtained, and this information, along with the matching result, can be sent to the target terminal. As an example, the material tag analysis information could include: adding content with unmatched material tags to the target material tag combination, or deleting content corresponding to material tags that do not match the target material tag combination, to improve the matching degree between the produced material and the target material tag combination. Returning different information to the target terminal based on different matching results provides accurate matching test information, further aiding in material production, reducing production costs, and saving resources.

[0147] In one example, the matching results can be displayed as follows: Figure 7 The target terminal can display the matching results of the target material on the matching results page, such as high matching degree, low matching degree, and material tag analysis information. This material tag analysis information can be as follows: Figure 7 The text indicates that "content can be added to unmatched material tags." Optionally, as shown... Figure 7 As shown, the matching results page can also display the matching and non-matching material tags for each target material, and can display various operation buttons, such as... Figure 7The options shown include "Select All," "Batch Delete," "Push to Video Library," "View Details," and "Delete." "Select All" selects all target materials on the matching results page; "Batch Delete" deletes all target materials on the matching results page; and "Push to Video Library," adjacent to "Select All," pushes all target materials from the matching results page to the video library. "View Details" and "Delete" allow users to view the details of the matching results for the corresponding target material and delete the corresponding target material. Match test details may include match score information, matching material tags, and non-matching material tags. Optionally, redundant material tags may also be included, such as material tags in the material tag information that do not match the target material tag combination. Figure 7 Redundant material labels are not shown.

[0148] As an application example, see Figure 8 The interactive illustration shown is a combination of recommended material tags. Figure 8 As shown, the content creator can trigger entry on the target terminal side. Figure 4a The shown content tag combination recommendation page can generate a content tag recommendation request. In response to this request, the content tag combination recommendation page is displayed, showing a list of content tag combinations. This list is obtained by a tag combination recommendation model that predicts the combinations of content tags and display association information for multiple historical content. The display metrics for the content tag combinations in this list meet the display metric conditions. The content tag combination list can be returned by the server in response to the content tag recommendation request; the specific method for obtaining it can be found in the relevant content above and will not be repeated here.

[0149] Furthermore, on the material tag combination recommendation page, users can also filter the material tag combinations in the list, for example, in... Figure 4a In the text, users can select filter criteria from multiple selection boxes, such as "scenario drama." A filter request can also be triggered. In response to a filter request for the list of material tag combinations, the display of sub-lists of material tag combinations can be toggled on the material tag combination recommendation page. Figure 4b As shown. This sublist of material tag combinations can be returned by the server in response to a filtering request for the material tag combination list. For details on how to obtain it, please refer to the relevant content above, which will not be repeated here.

[0150] Optionally, clicking any material tag combination in the material tag combination list can trigger a details request for the target material tag combination. Accordingly, in response to the details request for the target material tag combination, a target details page corresponding to the target material tag combination can be displayed. This target details page displays associated information about the target details page corresponding to the target material tag combination, such as... Figure 5 As shown. The target material tag combination is any combination of material tags from the material tag combination list.

[0151] In one example, the target details page can also display matching test controls, such as... Figure 5 The matching test is shown. Accordingly, the method may further include: in response to a matching test request triggered by a matching test control, displaying the matching result of the target material and the target material tag combination; the target material can be the material corresponding to the matching test request, i.e., the uploaded material. For example, a video can be uploaded to perform a matching test on the match between the video and the target material tag combination. For example, clicking the matching test can generate a matching test request, which can carry the target material tag combination and the target material. The matching test request can also be sent to the server, where the server performs the matching test and returns the matching result. The target terminal can display the matching result, such as... Figure 7 As shown. Furthermore, any target material on the matching test results page can be pushed to a material library. For example, the target material can be pushed to the material library, and materials in the material library can be recommended. This disclosure does not limit the recommendation processing method.

[0152] By providing the terminal side with a material tag combination recommendation page, a material tag combination details page, a matching test page, and a matching test result page, it can provide material creators with comprehensive and accurate recommendations and filtering of material tag combinations, as well as display of material tag combination matching results. It provides comprehensive and accurate reference information for material creation, saves material creation resources, achieves better material display effects, and can improve material creation efficiency and resource utilization.

[0153] Figure 9 This is a block diagram illustrating a material label recommendation device according to an exemplary embodiment. (Refer to...) Figure 9 The device may include:

[0154] The material tag combination list recommendation module 901 is configured to execute a material tag recommendation request in response to a target terminal, obtain and send a material tag combination list to the target terminal. The material tag combination list is obtained by performing tag combination prediction processing on the material tags and display association information corresponding to multiple historical materials based on a tag combination recommendation model. The display indicator information corresponding to the material tag combinations in the material tag combination list meets the display indicator conditions.

[0155] The material tag combination list filtering module 903 is configured to perform a filtering request on the material tag combination list and extract target object information from the filtering request. The target object information is used to indicate the association information of the target object to which the material is pushed.

[0156] The sublist acquisition module 905 is configured to retrieve a sublist of material tag combinations that match the target object information from the material tag combination list;

[0157] The sublist recommendation module 907 is configured to send the material tag combination sublist to the target terminal.

[0158] In one possible implementation, the device may further include:

[0159] The acquisition module is configured to periodically acquire multiple historical materials and their corresponding display association information.

[0160] The first prediction module is configured to perform material tag prediction processing by inputting the multiple historical materials into the material tag prediction model to obtain the material tags corresponding to each historical material.

[0161] The second prediction module is configured to input the material identifiers corresponding to each historical material, the display association information, and the material tags into the tag combination recommendation model, perform tag combination recommendation processing, and obtain the material tag combination list.

[0162] Among them, the material tags corresponding to each historical material are at least one of the preset material tag sets, and the material tags in the material tag combination are multiple of the preset material tag sets; the preset material tags in the preset material tag set represent the information of the material in multiple production dimensions.

[0163] In one possible implementation, the device may further include:

[0164] The details response module is configured to receive a details request from the target terminal for a target material tag combination, and obtain the target details page association information corresponding to the target material tag combination; the target material tag combination is any material tag combination in the material tag combination list;

[0165] The sending module is configured to send the information associated with the target details page to the target terminal.

[0166] In one possible implementation, the device may further include:

[0167] The matching test response module is configured to execute a matching test request in response to the target details page and obtain the target material to be matched;

[0168] The third prediction module is configured to perform prediction processing on the material tags of the target material based on the material tag prediction model, so as to obtain the material tag information corresponding to the target material;

[0169] The matching test module is configured to perform matching processing on the target material tag combination and the material tag information to obtain the matching result of the target material and the target material tag combination.

[0170] In one possible implementation, the matching test module may include:

[0171] The tag determination unit is configured to determine the first tag that is the same as the target material tag combination in the material tag information and the second tag after merging and deduplication;

[0172] The similarity information determination unit is configured to determine the similarity information between the target material tag combination and the material tag information based on the first weighted information of the first tag and the second weighted information of the second tag;

[0173] The matching result acquisition unit is configured to acquire the matching result based on the similarity information.

[0174] In one possible implementation, the device may further include:

[0175] The first matching processing module is configured to send the matching result to the target terminal and push the target material to the material library when the matching result meets the matching conditions.

[0176] The second matching processing module is configured to, when the matching result does not meet the matching conditions, obtain material tag analysis information and send the material tag analysis information and the matching result to the target terminal.

[0177] In one possible implementation, the details response module may include:

[0178] The description information and reference material acquisition unit is configured to acquire the detailed description information corresponding to each material tag in the target material tag combination, as well as the reference material corresponding to the target material tag combination;

[0179] The details association information acquisition unit is configured to acquire the target details page association information corresponding to the target material tag combination based on the detailed description information and the reference material.

[0180] In one possible implementation, the material tag combination list recommendation module 901 may include:

[0181] The tag recommendation response unit is configured to execute a material tag recommendation request in response to the target terminal, and obtain the material tag combination list, the preset object information corresponding to each material tag combination in the material tag combination list, the recommendation attribute information, and the corresponding reference video list;

[0182] The recommendation information generation unit is configured to generate material tag combination recommendation information based on the material tag combination list, the preset object information, the recommendation attribute information, and the reference video list.

[0183] The recommendation information sending unit is configured to send the material tag recommendation information to the target terminal.

[0184] This disclosure also provides a material tag recommendation device, applied to a terminal, which may include:

[0185] The recommendation page display module is configured to respond to material tag recommendation requests and display a material tag combination recommendation page. The material tag combination recommendation page displays a list of material tag combinations. The list of material tag combinations is obtained by a tag combination recommendation model through tag combination prediction processing of the material tags and display association information corresponding to multiple historical materials. The display indicator information corresponding to the material tag combinations in the material tag combination list meets the display indicator conditions.

[0186] The recommended display switching module is configured to switch the display of a sub-list of material tag combinations on the material tag combination recommendation page in response to a filtering request for the material tag combination list.

[0187] In one possible implementation, the device may further include:

[0188] The details page display module is configured to respond to a details request for a target material tag combination and display the target details page corresponding to the target material tag combination. The target details page displays the target details page association information corresponding to the target material tag combination. The target material tag combination is any material tag combination in the material tag combination list.

[0189] In one possible implementation, the target details page further displays a matching test control, and the device may further include:

[0190] The matching test module is configured to execute a matching test request triggered based on the matching test control, and display the matching result of the target material and the target material tag combination; the target material is the material corresponding to the matching test request.

[0191] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0192] Figure 10 This is a block diagram illustrating an electronic device for recommending material tags according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the electronic device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for recommending material tags. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0193] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0194] Figure 11This is a block diagram illustrating an electronic device for recommending material tags according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 11 As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for recommending material tags.

[0195] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0196] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the material tag recommendation method as described in the embodiments of this disclosure.

[0197] In an exemplary embodiment, a computer-readable storage medium is also provided, which, when executed by a processor of an electronic device, enables the electronic device to perform the material tag recommendation method of this disclosure. The computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0198] In an exemplary embodiment, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the material tag recommendation method of the embodiments of this disclosure.

[0199] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0200] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0201] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for recommending material tags, characterized in that, include: In response to a material tag recommendation request from a target terminal, a list of material tag combinations is acquired and sent to the target terminal. This list is obtained by a tag combination recommendation model that performs tag combination prediction processing on the material tags and display association information corresponding to multiple historical materials. The display indicator information corresponding to the material tag combinations in the list meets the display indicator conditions. The tag combination recommendation model is obtained by embedding vector learning and iterative learning on a preset deep learning model based on the material identifier corresponding to the sample material, the sample display association information, and the corresponding material tags. Each historical material corresponds to at least one material tag from a preset material tag set, and the material tags in the material tag combination are multiple material tags from the preset material tag set; the preset material tags in the preset material tag set represent information about the material in multiple production dimensions; In response to a filtering request for the list of material tag combinations, target object information is extracted from the filtering request. The target object information is used to indicate the associated information of the target object to which the material is pushed. Obtain a sub-list of material tag combinations that match the target object information from the material tag combination list; The list of material tag combinations is sent to the target terminal.

2. The method according to claim 1, characterized in that, The method further includes: Periodically retrieve multiple historical materials and their corresponding display information; The multiple historical materials are input into the material tag prediction model for material tag prediction processing to obtain the material tags corresponding to each historical material. The material identifiers corresponding to each historical material, the display association information, and the material tags are input into the tag combination recommendation model to perform tag combination recommendation processing, thereby obtaining the material tag combination list.

3. The method according to claim 1 or 2, characterized in that, The method further includes: Upon receiving a details request from the target terminal for a target material tag combination, the system obtains the target details page association information corresponding to the target material tag combination; the target material tag combination is any material tag combination in the material tag combination list. Send the associated information of the target details page to the target terminal.

4. The method according to claim 3, characterized in that, The method further includes: In response to a matching test request from the target details page, obtain the target material to be matched; Based on the material tag prediction model, the material tags of the target material are predicted to obtain the material tag information corresponding to the target material; The target material tag combination is matched with the material tag information to obtain the matching result of the target material and the target material tag combination.

5. The method according to claim 4, characterized in that, The process of matching the target material tag combination with the material tag information to obtain the matching result of the target material and the target material tag combination includes: Identify the first tag that is the same as the target material tag combination in the material tag information, and the second tag after merging and deduplication; Based on the first weighted information of the first tag and the second weighted information of the second tag, the similarity information between the target material tag combination and the material tag information is determined; The matching result is obtained based on the similarity information.

6. The method according to claim 4 or 5, characterized in that, The method further includes: If the matching result meets the matching conditions, the matching result is sent to the target terminal, and the target material is pushed to the material library; If the matching result does not meet the matching conditions, obtain the material tag analysis information and send the material tag analysis information and the matching result to the target terminal.

7. The method according to claim 3, characterized in that, The step of obtaining the target details page association information corresponding to the target material tag combination includes: Obtain detailed descriptions of each material tag in the target material tag combination, as well as reference materials corresponding to the target material tag combination; Based on the detailed description and the reference materials, obtain the target details page association information corresponding to the target material tag combination.

8. The method according to claim 1 or 2, characterized in that, The step of responding to a material tag recommendation request from a target terminal by obtaining and sending a list of material tag combinations to the target terminal includes: In response to the material tag recommendation request from the target terminal, the material tag combination list, the preset object information corresponding to each material tag combination in the material tag combination list, the recommendation attribute information, and the corresponding reference video list are obtained; Based on the material tag combination list, the preset object information, the recommendation attribute information, and the reference video list, generate material tag combination recommendation information; The recommended information for the material tags is sent to the target terminal.

9. A method for recommending material tags, characterized in that, include: In response to a material tag recommendation request, a material tag combination recommendation page is displayed. This page shows a list of material tag combinations, which is obtained by a tag combination recommendation model that predicts combinations of material tags and display association information for multiple historical materials. The display metrics information corresponding to the material tag combinations in the list meets the display metric conditions. The tag combination recommendation model is obtained by embedding vector learning and iterative learning on a preset deep learning model based on the material identifier, display association information, and corresponding material tags of the sample material. Each historical material corresponds to at least one material tag from a preset material tag set, and the material tags in the material tag combination are multiple material tags from the preset material tag set; the preset material tags in the preset material tag set represent information about the material in multiple production dimensions; In response to a filtering request for the list of material tag combinations, the sub-list of material tag combinations is switched to be displayed on the material tag combination recommendation page.

10. The method according to claim 9, characterized in that, The method further includes: In response to a request for details of a target material tag combination, a target details page corresponding to the target material tag combination is displayed. The target details page displays the associated information of the target details page corresponding to the target material tag combination. The target material tag combination is any material tag combination in the list of material tag combinations.

11. The method according to claim 10, characterized in that, The target details page also displays a matching test control, and the method further includes: In response to a matching test request triggered by the matching test control, the matching result of the target material and the target material tag combination is displayed; the target material is the material corresponding to the matching test request.

12. A material label recommendation device, characterized in that, include: The material tag combination list recommendation module is configured to execute a material tag recommendation request in response to a target terminal, obtain and send a material tag combination list to the target terminal. The material tag combination list is obtained by a tag combination recommendation model based on the material tags and display association information corresponding to multiple historical materials, and the display indicator information corresponding to the material tag combinations in the material tag combination list meets the display indicator conditions. The tag combination recommendation model is obtained by embedding vector learning and iterative learning on a preset deep learning model based on the material identifier corresponding to the sample material, the sample display association information, and the corresponding material tags. Each historical material corresponds to at least one material tag from a preset material tag set, and the material tags in the material tag combination are multiple material tags from the preset material tag set; the preset material tags in the preset material tag set represent information about the material in multiple production dimensions; The material tag combination list filtering module is configured to perform a filtering request in response to the material tag combination list, extracting target object information from the filtering request, wherein the target object information is used to indicate the association information of the target object to which the material is pushed; The sublist retrieval module is configured to retrieve a sublist of material tag combinations that match the target object information from the material tag combination list; The sublist recommendation module is configured to send the sublist of material tags to the target terminal.

13. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the material tag recommendation method as described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of an electronic device, the electronic device is enabled to perform the material label recommendation method as described in any one of claims 1 to 11.

15. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the material tag recommendation method according to any one of claims 1 to 11.

Images